Monday 20th of November 2017 12:54:23 AM MySQL Stored Procedures: Part 1
August 9, 2005

What are Stored Procedures

MySQL 5.0 finally introduces functionality for Stored Procedures. So what exactly are stored procedures? That is the kind of question that gets database professionals who use other DBMS's raising their eyebrows. Stored procedures have been integral to Oracle, PostgreSQL, DB-2, MS-SQL server and others for years, and it has long been a sore point that MySQL has not had them. But there is no snobbery here - if you are a MySQL newbie, or have been using MySQL for years and want to find out what all the fuss is about, read on. If it is your eyebrows that are raised, and you just want to know how MySQL implements them, you will be relieved to know MySQL stored procedures are very similar to the DB2 implementation, as both are based on the SQL:2003 standard.

A stored procedure is simply a procedure that is stored on the database server. MySQL developers have to date unthinkingly written and stored their procedures on the application (or web) server, mainly because there hasn't been an option. That has been limiting. Some have claimed that there are two schools of thought - one claiming that logic should be in the application, the other saying it should reside in the database. However, most professionals would not bind themselves to one or other viewpoint at all times. As always, there are times when doing either makes sense. Unfortunately, some of the staunchest adherents of the in the application school are only there because until now they have had no choice, and it is what they are used to doing. So why would we want to place logic on the database server?

Why use stored procedures?

A simple example

A stored procedure is simply some SQL statements. Almost any valid SQL can go inside a stored procedure, with a few exceptions, which we will look at, at a later date. Let's set up a basic stored procedure first. This one will simply say 'Hello' in the Xhosa language - Molo.

mysql> CREATE PROCEDURE molo() SELECT 'Molo';
Query OK, 0 rows affected (0.00 sec)

It is as simple as that. And to call it:

mysql> CALL molo()\G
*************************** 1. row ***************************
Molo: Molo
1 row in set (0.00 sec)

Hardly useful, but the basics are there. CREATE PROCEDURE sp_name() will define the procedure, and CALL sp_name() will call the procedure.

Parameters

The real benefit of a stored procedure is of course when you can pass values to it, as well as receive values back. The concept of parameters should be familiar to anyone who has had experience with any procedural programming experience.

There are three types of parameter:

Mastery of stored procedures does require knowledge of session variables. Most of you probably know how to use session variables already, but if not, the concept is simple. You can assign a value to a variable, and retrieve it later. Here is an example, setting the variable x to the Xhosa word for hello to a group of people.

mysql> SET @x='Molweni';
Query OK, 0 rows affected (0.00 sec)
mysql> SELECT @x\G
*************************** 1. row ***************************
@x: Molweni
1 row in set (0.00 sec)

An IN example

Here is an example of a stored procedure demonstrating the use of an IN parameter. Since IN is the default, there is no need to specify the parameter as such.

mysql> CREATE PROCEDURE sp_in(p VARCHAR(10)) SET @x = P;
Query OK, 0 rows affected (0.00 sec)
mysql> CALL sp_in('Molo');
Query OK, 0 rows affected (0.00 sec)
mysql> SELECT @x\G
*************************** 1. row ***************************
@x: Molo
1 row in set (0.00 sec)

The session variable @x is set inside of the procedure, based upon the parameter P, which is passed to the procedure, and remains unchanged.

An OUT example

mysql> SET @x='Molweni';
Query OK, 0 rows affected (0.00 sec)

Speaking of list items, they have a few special rules in addition to everything discussed so far. List items are typically preceded by a marker, such as a small dot or a number. This marker isn't actually part of the list-item's content area, so you get effects like those illustrated in Figure 8-25. mysql> CREATE PROCEDURE sp_out(OUT p VARCHAR(10)) SET P='molo'; Query OK, 0 rows affected (0.00 sec) mysql> CALL sp_out(@x); Query OK, 0 rows affected (0.00 sec) mysql> SELECT @x\G *************************** 1. row *************************** @x: molo 1 row in set (0.00 sec)

We reset @x just to make sure the final result is not a legacy of the previous procedure. This time, the parameter P is changed inside of the procedure, while the session variable is passed to the procedure, ready to receive the result.

An INOUT example

mysql> CREATE PROCEDURE sp_inout(INOUT P INT) SET @x=P*2;
Query OK, 0 rows affected (0.00 sec)
mysql> CALL sp_inout(2);
Query OK, 0 rows affected (0.00 sec)
mysql> SELECT @x\G
*************************** 1. row ***************************
@x: 4
1 row in set (0.00 sec)

Here, a parameter is passed to the procedure, used in the calculation, and the results are made available to the session variable @x.

child paragraph. The next element after the DIV is the appropriate distance from the bottom of the DIV. The fact that it overlaps the paragraph doesn't matter, at least not technically.

Now let's consider an example where the margins of a list item, an unordered list, and a heading are all collapsed. In this case, the unordered list and heading will be set to have negative margins:

LI {margin-bottom: 20px;}
variants might be assigned values for font-weightas shown in Table 5-3.

Table 5-1. Hypothetical Weight Assignments for a Specific Font

The first three number values are assigned to the lightest weight.The Regular face gets the keywords 400 andnormal, as expected. Since there is a Medium font,it's assigned to the number 500. There isnothing to assign to 600, so it's mapped tothe Bold font face, which is also the variant to which

  • For nonroot elements that are absolutely positioned using aposition of absolute, thecontaining block is set to the nearest ancestor (of any kind) thathas a position other thanstatic. This happens as follows:

    1. If the ancestor is block-level, the containing block is set to bethat element's padding edge; in other words, the area that